← Back to changelog

January 31, 2025

OpenAI o3-mini support for playground and cost tracking

Marc Klingen

Two hours ago, OpenAI released the latest version of their o3-mini model. Langfuse now supports this model in both the LLM playground and cost tracking.

Some highlights about o3-mini from the OpenAI release notes:

• Performance: Matches o1 in math, coding, and science; responds 24% faster.
• Developer Features: Supports function calling, structured outputs, and developer messages.
• Reasoning Effort Levels: Offers low, medium, and high options to balance speed and thoroughness.

Learn more

• Langfuse LLM playground
• Langfuse model usage and cost tracking